This study analyzed the risk factors in patients who developed distal junctional kyphosis (DJK) after posterior cervical fusion.
We retrospectively analyzed the clinical and radiographic outcomes of 64 patients, aged ≥18 years (51 and 13 male and female patients, respectively), who underwent single-staged multilevel (3–6 levels) posterior cervical fusion surgery due to multiple cervical spondylotic myelopathy. The surgeries were performed by a single spinal surgeon between January 2012 and December 2017. Demographic data, clinical outcomes, and radiological results were collected. We divided the patients into a DJK group and a non-DJK group according to the presence of DJK and investigated the risk factors by comparing the differences between the two groups.
Of the 64 patients, 13 developed DJK. No significant differences in clinical results were observed between the two groups before and immediately after the surgery. At the final follow-up, a higher visual analog score for neck pain was observed in the DJK group compared to the non-DJK group (
DJK can be considered to be caused by cervical misalignment due to excessive change in the surgical site in patients with low T1 slope and high C2-7 SVA before surgery. This also affects the clinical outcome after surgery. It is recommended to refrain from excessive segmental lordosis changes during multilevel cervical post fusion surgery, especially in patients with a small preoperative T1 slope and a large SVA value.
Cervical spondylotic myelopathy (CSM) is one of the most common spinal disorders treated by spinal surgeons worldwide. CSM often presents in the elderly with several signs and symptoms, including neck pain, radiculopathy, or myelopathy [
Posterior cervical spinal fusion with decompression is one of the possible treatment options in patients with CSM who have conditions such as spondylosis, spinal stenosis, and degenerative disc disease [
DJK may lead to fixation failure, disc degeneration, distal level fractures, or spondylolisthesis, resulting in pain, myelopathy, or deformities [
Approval for this study was obtained from the participating center’s Institutional Review Board of Gangneung Asan Hospital (IRB No. GNAH 2020-12-009). We retrospectively analyzed the clinical and radiographic outcomes of 64 patients aged ≥18 years (51 males and 13 females) who had multilevel CSM and underwent single-stage multilevel (3–6 levels) posterior cervical fusion surgery. The surgeries were performed by a single spine surgeon between January 2012 and December 2017. The surgical level was from C2 to C7. The Neck Disability Index (NDI) score and Visual analog score (VAS) of the arm and neck were used to measure clinical results. We also compared preoperative and postoperative (last follow-up) values.
Demographic data were collected for all patients, including age, sex, underlying disease, smoking history, body mass index (BMI), surgical details, follow-up information, electronic medical records, and radiographic image reviews. Exclusion criteria were previous cervical spine surgery, anterior surgery, follow-up less than 24 months, and surgery for other reasons.
We divided patients into DJK and non-DJK groups according to the presence of DJK at the last follow-up. DJK was defined as the changes of an angle of -10° or less at the distal disc level from the end of the fusion construct between baseline and final follow up [
We performed cervical pedicle screw (CPS) insertion in all posterior cervical spinal fusion patients. The safety and efficacy of subaxial CPS placement have been validated multiple times [
Following screw insertion and decompression, we extended the patient’s head position using remote-controlled table head segments to achieve ideal lordosis from 15–30 degrees [
Cervical lordosis (CL) from C2 to C7, the T1 slope, the segmental angle index level, and the C2-7 sagittal vertical axis (SVA) were evaluated via lateral cervical X-ray (
All radiographic parameters were compared between the two groups.
Data are expressed as percentages and presented as mean± standard deviation. Student’s t-test was used to analyze the differences between continuous variables.
A total of 64 patients with multilevel CSM were included in this study, where 13 developed DJK and 51 did not. Of the 13 patients with DJK, three were asymptomatic and 10 experienced neck pain around the operated site (VAS score ≥4) after surgery (
The mean patient age was 60.85±10.18 years, and the mean follow-up period was 35.09±11.13 months. No significant difference between DJK and non-DJK groups was observed in sex, age, underlying diseases, BMI, or smoking history (
The clinical outcomes of all the patients showed improvement. There were no significant differences in the VAS scores of the arm or neck or in the NDI between the two groups before and after surgery. Higher NDI and VAS values were observed at the last follow-up in the DJK group than in the non-DJK group. A significant difference was observed in the VAS scores of the neck (
Significant differences were found in the T1 slope and C2-7 SVA (
In all the patients, the cervical spinal radiological parameters tended to return to their preoperative values (
When examining the changes between radiological parameters before surgery, immediately after surgery, and at the last follow-up, the changes in T1 slope, CL, and segmental angle before and immediately after surgery were significantly larger in the DJK group (
The changes in CL, and T1-Cl mismatch between immediately after surgery and the last follow-up were significantly larger in the DJK group (
From the multiple logistic regression analysis of DJK occurrence, preoperative higher C2-7 SVA and preoperative lower T1 slope were identified as independent risk factors (
We investigated the risk factors for DJK development after posterior decompression and fusion surgery. The T1 slope and CL displayed a tendency to return to their preoperative levels. Although many studies have been published on the relationship between the T1 slope and CL, to date [
The return of the T1 slope and CL to preoperative levels may have resulted from patients tending to retain their preoperative spinopelvic alignments, which is due to the preservation of the preoperative T1 slope [
The multiple logistic regression analysis in this study showed that the preoperative C2-7 SVA and T1 slope are independent risk factors for DJK (
These findings are similar to those in the lumbar deformity literature on the incidence of proximal junction kyphosis. Kim and Iyer [
Neither age, sex, BMI, underlying disease, nor smoking significantly increased the risk of DJK. Although many studies have mentioned the relationship between smoking and cervical disc degeneration [
We found in this study that four of the 28 patients (14%) completed the post-fusion surgery at C6 and nine of the 31 patients (29%) ended up at C7. It has been found that DJK occurs at the cervicothoracic junction (
In 2018, Passias et al. [
There are some limitations to this study. First, this was a retrospective study. Second, the patients were examined only with X-ray imaging and not with computed tomography or magnetic resonance imaging. Finally, these results were derived from a small sample of patients. Studies with larger cohorts are needed. In addition, many patients who showed improvement were excluded because they did not meet the follow-up requirement, which may have resulted in selection bias.
In our study, risk factors of DJK in patients who developed DJK after posterior cervical fusion were analyzed. In patients with lower T1 slope and large C2-7 SVA, misalignment occurred due to excessive segment angle change during surgery, eventually resulting in DJK at last follow-up. In the clinical results, it was confirmed that the clinical results deteriorated again in the DJK group at the last follow-up. It is recommended to refrain from excessive SL changes during multilevel cervical post fusion surgery, especially in patients with a small preoperative T1 slope and a large SVA value.
No potential conflict of interest relevant to this article was reported.
This type of study does not require informed consent.
Conceptualization : JHP; Data curation : JJL; Formal analysis : JJL, YGO; Methodology : HKS, BGP; Project administration : JHP; Visualization : YGO, BGP; Writing - original draft : JJL; Writing - review & editing : JHP
None
None
Standard lateral cervical X-ray before and immediately after surgery. Standard lateral cervical X-ray before (A) and immediately after surgery (B). Pre-op : preoperative, SL : segmental lordosis, CL : cervical lordosis, SVA : sagittal vertical axis, Post-op : postoperative.
T1 slope, cervical lordosis, segmental lordosis and sagittal vertical axis values preoperatively, postoperatively, and at the final follow-up. Final follow-up and preoperative values for T1 slope, cervical lordosis, and sagittal vertical axis were similar. These similarities were statistically significant (
CL, SL, T1 slope increased and C2-7 SVA decreased immediately but eventually returned to preoperative values (A). Changes to T1 slope, CL, SL, and C2-7 SVA immediately postoperatively (B) and at the final follow-up (C). CL : cervical lordosis, SL : segmental lordosis, SVA : sagittal vertical axis.
Algorithm of DJK development. SVA : sagittal vertical axis, SL : segmental lordosis, CL : cervical lordosis, f/u : follow-up, DJK : distal junctional kyphosis.
Characteristics of patients who developed DJK after surgery
No. | Age (years)/sex | Fusion level | Follow-up duration (months) | Revision | Symptom at last follow-up |
---|---|---|---|---|---|
1 | 55/M | C3-4-5-6-7 | 27 | X | Posterior neck pain |
2 | 70/F | C4-5-6-7 | 23 | X | Asymptomatic |
3 | 76/M | C3-4-5-6-7 | 24 | X | Posterior neck pain |
4 | 64/M | C3-4-5-6-7 | 30 | X | Posterior neck pain |
5 | 55/M | C3-4-5-6 | 38 | X | Posterior neck pain |
6 | 63/F | C3-4-5-6 | 24 | X | Posterior neck pain |
7 | 47/M | C4-5-6-7 | 24 | X | Posterior neck pain |
8 | 80/F | C4-5-6 | 30 | X | Asymptomatic |
9 | 55/M | C5-6-7 | 32 | X | Posterior neck pain |
10 | 73/M | C3-4-5-6 | 36 | ACDF C6-7 | Posterior neck pain and radiculopathy |
11 | 57/M | C5-6-7 | 26 | X | Posterior neck pain |
12 | 69/M | C4-5-6-7 | 25 | X | Asymptomatic |
13 | 55/M | C3-4-5-6-7 | 24 | X | Posterior neck pain |
DJK : distal junctional kyphosis, M : male, F : female
Demographics of the DJK and non-DJK groups
DJK group (n=13) | Non-DJK group (n=51) | ||
---|---|---|---|
Age (years) | 62.15±9.50 | 60.52±10.41 | 0.61 |
Sex, M : F | 8 : 5 | 45 : 6 | 0.28 |
Fusion level | 3.30±0.85 | 2.84±0.88 | 0.09 |
Hypertension | 6 | 17 | 0.39 |
Diabetes mellitus | 4 | 11 | 0.43 |
Smoking | 6 | 19 | 0.19 |
Body mass index (kg/m2) | 24.19±2.66 | 24.47±3.56 | 0.79 |
Follow-up duration (months) | 39.31±12.44 | 28.83±12.14 | 0.14 |
Values are presented as mean±standard deviation or number (%) unless otherwise indicated. DJK : distal junctional kyphosis, M : male, F : female
Clinical outcomes for each period in both groups
DJK group (n=13) | Non-DJK group (n=51) | ||
---|---|---|---|
VAS, neck | |||
Preoperative | 7.69±1.31 | 7.11±2.02 | 0.33 |
Postoperative | 2.23±1.96 | 3.41±2.25 | 0.08 |
Last f/u | 2.63±1.75 | 1.62±1.19 | 0.01 |
VAS, arm | |||
Preoperative | 5.53±2.75 | 6.72±2.20 | 0.10 |
Postoperative | 1.92±0.86 | 2.76±1.64 | 0.09 |
Last f/u | 2.61±1.60 | 2.31±1.71 | 0.56 |
NDI | |||
Preoperative | 26.76±11.05 | 26.50±10.77 | 0.93 |
Postoperative | 10.07±7.34 | 13.43±9.51 | 0.24 |
Last f/u | 15.61±8.37 | 10.47±8.45 | 0.06 |
Values are presented as mean±standard deviation. DJK : distal junctional kyphosis, VAS : visual analog scale, f/u : follow-up, NDI : Neck Disability Index
Comparison of radiologic parameters between the DJK and non-DJK groups
DJK group (n=13) | Non-DJK group (n=51) | ||
---|---|---|---|
T1 slope (°) | |||
Preoperative | 20.26±6.53 | 25.64±8.57 | 0.03 |
Immediate postoperative | 28.94±7.53 | 29.03±6.67 | 0.96 |
Last f/u | 24.20±8.31 | 26.41±8.31 | 0.36 |
CL (°) | |||
Preoperative | 8.07±9.07 | 11.21±9.07 | 0.29 |
Immediate postoperative | 19.38±11.27 | 17.01±8.62 | 0.41 |
Last f/u | 9.84±9.71 | 13.23±8.63 | 0.22 |
SL (°) | |||
Preoperative | 4.69±7.75 | 7.11±7.57 | 0.30 |
Immediate postoperative | 15.15±11.31 | 11.96±6.40 | 0.18 |
Last f/u | 11.38±10.68 | 9.58±6.40 | 0.43 |
C2-7 SVA (mm) | |||
Preoperative | 31.92±9.24 | 22.29±8.95 | <0.01 |
Immediate postoperative | 22.13±8.54 | 16.96±9.24 | 0.08 |
Last f/u | 30.59±13.88 | 21.57±8.86 | <0.01 |
T1 slope – CL(°) | |||
Preoperative | 12.18±4.40 | 14.42±6.68 | 0.25 |
Immediate postoperative | 9.56±6.45 | 12.02±7.53 | 0.28 |
Last f/u | 16.63±5.72 | 13.18±7.79 | 0.14 |
Values are presented as mean±standard deviation. DJK : distal junctional kyphosis, f/u : follow-up, CL : cervical lordosis, SL : segmental lordosis, SVA : sagittal vertical axis
Differences in cervical parameters before and after surgery
DJK group (n=16) | Non-DJK group (n=25) | ||
---|---|---|---|
Δ T1 slope | 8.69±4.04 | 3.40±5.81 | 0.03 |
Δ CL | 11.30±7.34 | 5.80±7.25 | 0.01 |
Δ SL | 10.46±7.95 | 4.84±6.04 | <0.01 |
Δ C2-7 SVA | 8.94±5.32 | 5.34±6.80 | 0.08 |
Δ T1 – CL | 2.61±4.51 | 2.40±6.82 | 0.91 |
Values are presented as mean±standard deviation. Δ = immediate postoperative value – preoperative value. DJK : distal junctional kyphosis, CL : cervical lordosis, SL : segmental lordosis, SVA : sagittal vertical axis
Differences in cervical parameters immediately postoperatively and at the final follow-up
DJK group (n=16) | Non-DJK group (n=25) | ||
---|---|---|---|
Δ T1 slope | 4.74±3.11 | 2.62±3.96 | 0.07 |
Δ CL | 9.53±5.39 | 3.78±4.77 | <0.01 |
Δ SL | 3.76±2.91 | 2.37±2.51 | 0.08 |
ΔC2-7 SVA | 7.22±8.18 | 4.74±5.32 | 0.10 |
Δ T1 – CL | 7.06±7.78 | 1.15±4.60 | <0.01 |
Values are presented as mean±standard deviation. Δ = final follow-up value − immediate postoperative value. DJK : distal junctional kyphosis, CL : cervical lordosis, SL : segmental lordosis, SVA : sagittal vertical axis
Multivariable stepwise logistic regression analysis results for factors influencing DJK after posterior cervical spinal fusion surgery
Risk factor | OR | 95% CI | |
---|---|---|---|
Preoperative T1 slope | 0.49 | 0.29–0.81 | <0.01 |
Preoperative SVA | 1.42 | 1.13–1.78 | 0.03 |
DJK : distal junctional kyphosis, OR : odds ratio, CI : confidence interval, SVA : sagittal vertical axis